Volume 27, Issue 1, Pages 167-179.e7 (January 2018) 17β-Estradiol Directly Lowers Mitochondrial Membrane Microviscosity and Improves Bioenergetic Function in Skeletal Muscle  Maria J. Torres, Kim A. Kew, Terence E. Ryan, Edward Ross Pennington, Chien-Te Lin, Katherine A. Buddo, Amy M. Fix, Cheryl A. Smith, Laura A. Gilliam, Sira Karvinen, Dawn A. Lowe, Espen E. Spangenburg, Tonya N. Zeczycki, Saame Raza Shaikh, P. Darrell Neufer  Cell Metabolism  Volume 27, Issue 1, Pages 167-179.e7 (January 2018) DOI: 10.1016/j.cmet.2017.10.003 Copyright © 2017 Elsevier Inc. Terms and Conditions

Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 1 2-Week-OVX Decreases Mitochondrial Function and Induces an Oxidative Shift in SM Cellular Redox Environment (A) Citrate synthase activity in red gastrocnemius (RG). (B) JO2 measured in PmFbs from RG. (C) Fatty acid-supported JO2. G/M, glutamate/malate; Succ, succinate; Rot, rotenone; AmA, antimycin A; Asc, ascorbate; TMPD, N,N,N′,N′-tetramethyl-p-phenylenediamine dihydrochloride; PC, palmitoyl-carnitine; P-CoA, almitoyl-CoA; L-Carn, L-carnitine. (D–F) Mitochondrial respiratory kinetics. Pyruvate titrations in the presence of malate and ADP (D), and ADP titrations in the presence of G/M (E) or pyruvate/malate (F). Kinetic parameters KM and Vmax were determined from fitting to Michaelis-Menten functions. Changes in Vmax were significant (∗p < 0.05). (G and H) Total GSH and GSSG concentration (G) and resulting GSH/GSSG ratios (H) in RG. (I) JH2O2 measured in PmFbs pre-incubated with or without 1-chloro-2,4-dinitrobenzene (CDNB) for GSH depletion, then added pyruvate. Values are means ± SEM; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001, n = 8–15 mice/group. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 2 E2 Therapy Reverses the OVX-Induced Pro-diabetogenic State (A) Study design. (B) Uterine mass at sacrifice. (C) Body composition. (D) Fasting blood glucose. !p < 0.05, !!p < 0.01 versus pre-OVX for each group, and #p < 0.05 versus OVX-4w[ctl]. (E) Glucose tolerance test. Inset: area under the curve (AUC). (F) Insulin tolerance test. Inset: slope of the least-squared-regression line calculated from the first 20 min. (G) HOMA-IR scores calculated as in Figure S2E. (H and I) Ex vivo basal and insulin-stimulated 3H-2-deoxy-glucose uptake in whole EDL (H) and soleus (I) muscles. Values are means ± SEM, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus NC-Pro and #p < 0.05, ##p < 0.01 versus OVX-4w[ctl], n = 6 mice/group. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 3 E2 Therapy Restores Mitochondrial Function and Cellular Redox Balance in SM (A) JO2 measured in PmFbs from RG. C I-supported JO2 measured by the sequential addition of pyruvate/malate (Pyr/Mal) and ADP (left panel); or glutamate/malate (G/M) and ADP (middle panel), and C II-supported JO2 measured in the presence of rotenone (Rot), succinate (Succ), and ADP (right panel). (B) Fatty-acid-supported JO2 in RG PmFbs, as in Figure 1C. (C) Total GSH and GSSG levels in whole gastrocnemius. (D) Resulting 2GSH/GSSG ratios. (E) Left: JH2O2 measured after the addition of Pyr, followed by 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU). !!p < 0.01 versus Pyr. Right: PmFBs were pre-incubated with 1-chloro-2,4-dinitrobenzene (CDNB), then added pyruvate. (F) JH2O2 measured after the addition of succinate, auranofin. (G) Mitochondrial free radical leak (JH2O2/JO2∗2) under succinate-supported respiration. Values are means ± SEM, ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 versus NC-Pro, and #p < 0.05, ##p < 0.01 versus OVX-4w[ctl], n = 6 mice/group. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 4 E2 Therapy Restores OVX-Induced Loss of C I, C I/II + III Activities (A and B) Western blot analysis of OXPHOS complexes (A) and citrate synthase activity in isolated mitochondria (B). (C) Relative specific activity of respiratory complexes normalized to citrate synthase activity (1 CSU = 1 μmol/min/mg protein). See Figure S4. Values are means ± SEM, ∗p < 0.05, ∗∗p < 0.01 versus NC-Pro, and #p < 0.05, ##p < 0.01 versus OVX-4w[ctl], n = 4–5 mice/group. NS, not significant. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 5 E2 Decreases Microviscosity in Biomimetic Mitochondrial Membranes (A) Sample fluorescence MC-540 emission spectra from LUVs made with PC:PE:PS and increasing [E2] or cholesterol (negative control). (B) Peak MC-540 F values from (A). Values are means ± SEM, from three independent LUVs preparations. ∗p < 0.05; ∗∗p < 0.01 versus [E2] 0%. (C) Representative pressure-area (π-A) isotherms of DPPC monolayers containing 5 mol % E2 or cholesterol. (D) Average mean molecular area (MmA) at relevant surface pressures extrapolated from three independent π-A isotherms. (E) The surface elasticity moduli (Cs−1) as a function of MmA, corresponding to the respective π-A isotherms shown in (C). (F) Cs−1 calculated from three independent π-A isotherms (dotted lines). Values are means ± SEM. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 6 E2 Localizes to Mitochondrial Membranes and Decreases Microviscosity Independently of ERα (A) Representative extracted ion chromatograms for E2. (B) E2 content in mitochondrial membrane extracts measured by LC-MS. (C) Fluorescence MC-540 emission spectra in intact SM mitochondria. (D) Peak MC-540 F values from C. Data are means ± SEM, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 versus NC-Pro, and #p < 0.05, ##p < 0.01 versus OVX-4w[ctl], n = 6–8 mice/group. (E and F) E2 content (E) and respective C I relative specific activity (F). Values are means ± SEM, ∗p < 0.05 versus NC-skmERαKO, and #p < 0.05 versus OVX-skmERαKO, n = 3–4 mice/group. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions

Figure 7 In Vitro E2 Exposure of OVX Mitochondria Restores Membrane Microviscosity, JH2O2 Emitting Potential and OXPHOS Steady-State Flux (A) Experimental design. (B) E2 content in mitochondrial membranes. Values are means ± SEM, n = 7–14 mice/group. (C) Peak F values from MC-540 emission spectra in fresh intact SM mitochondria pre-incubated ± 3 nM E2. Values are means ± SEM of three replicate measures, n = 8 mice/group. (D) Specific activities of complexes, measured as in Figure 4. n = 7 mice/group. (E and F) JH2O2 (E) and maximal JO2 capacity (F) measured in isolated mitochondria as in Figure 1B. (G–I) (G) JO2 (G), JATP (H), and resulting ATP/O ratios (JATP/JO2) (I), in the presence of G/M/Pyr/Succ and ADP. Values are means ± SEM. ∗p < 0.05, ∗∗p < 0.01 versus NC-Pro, and !p < 0.05, !!p < 0.01 versus OVX-4w. n = 7 mice/group. ns, not significant. Cell Metabolism 2018 27, 167-179.e7DOI: (10.1016/j.cmet.2017.10.003) Copyright © 2017 Elsevier Inc. Terms and Conditions